Chapter 1 — Quiz

Twenty questions. Answer each before opening the explanation — the attempt is what produces retention, even when the attempt fails.

Multiple Choice

1. Which of the following is the strongest reason a randomized controlled trial cannot settle whether red meat causes heart disease?

  • a) Red meat is too expensive to supply to a study population
  • b) Heart disease develops over decades, and no one has kept free-living humans on an assigned diet that long
  • c) Red meat consumption is illegal in some jurisdictions
  • d) Randomization is not possible for dietary exposures
Answer **b.** The timescale is the binding constraint. Randomization to a diet *is* possible (d is false) — it just can't be sustained for the decades that hard cardiovascular outcomes require, and dropout is non-random. Cost (a) matters but is not the fundamental barrier.

2. "The control group in a diet study is eating something too" is a statement of which problem?

  • a) Publication bias
  • b) The substitution question
  • c) Reverse causation
  • d) Social desirability bias
Answer **b.** There is no diet-naive human, so every dietary comparison is a substitution rather than a deletion. This is why "is butter bad?" has no answer until you specify what replaces it.

3. Food frequency questionnaire errors are described in the chapter as "systematic, not random." The practical consequence is that:

  • a) Larger sample sizes will eventually eliminate the error
  • b) The errors cancel out across a population
  • c) Larger sample sizes will not fix the problem, because the bias has a consistent direction
  • d) FFQs should never be used in research
Answer **c.** Random error shrinks with sample size; systematic error does not — you just get a more precise estimate of a biased quantity. (d) overshoots: FFQs remain useful, but their output is a hypothesis rather than a measurement.

4. In the five-step pipeline, which step does the chapter identify as doing the most damage, even though it contains no false statements?

  • a) The original study
  • b) The press release and the news article
  • c) The aggregation
  • d) The reel
Answer **b.** Steps 1 and 2 move the emphasis while remaining technically true — dropping "mice," "high dose," and "marker" from the headline. By the time an outright falsehood appears at step 3, the ground is prepared. Misleading-but-true survives fact-checking; false does not.

5. Which of these is described in the chapter as a genuine, substantial reversal of scientific guidance, representing the field working correctly?

  • a) The recommendation on eggs
  • b) The recommendation on infant allergen introduction
  • c) The recommendation on fiber
  • d) The recommendation on folic acid in pregnancy
Answer **b.** Guidance shifted from delaying allergenic foods in at-risk infants toward early introduction, on the basis of randomized evidence. That is a real reversal driven by better data. Eggs (a) shifted modestly in the evidence and dramatically in the reporting; fiber (c) and folate (d) have been stable.

6. The chapter argues that the food industry and the wellness industry are best understood as:

  • a) Adversaries, with the wellness industry acting as a check on the food industry
  • b) Independent, with no meaningful relationship
  • c) Symbiotic — fear of the first is the business model of the second
  • d) Identical, since they are owned by the same parent companies
Answer **c.** The wellness industry's acquisition channel is fear of the food industry's products. This is why wellness content runs angrier than food advertising: anger converts.

7. Which myth "shape" is at work in the claim "there's one ingredient in your pantry that's causing your fatigue, and the industry doesn't want you to know"?

  • a) The single cause and the suppressed truth
  • b) Restriction as virtue and the lost golden age
  • c) Purity and contamination only
  • d) The hidden simple fix and restriction as virtue
Answer **a.** "One ingredient" is the single cause; "doesn't want you to know" is the suppressed truth. The suppressed-truth shape is especially resilient because absence of supporting evidence gets reinterpreted as proof of the suppression.

8. According to the chapter, why is media coverage of nutrition anti-correlated with reliability?

  • a) Journalists are poorly educated in science
  • b) Coverage selects for novelty and reversal, which is precisely the subset of findings least likely to replicate
  • c) Journals refuse to share papers with the press
  • d) Most nutrition research is fraudulent
Answer **b.** The selection filter is newsworthiness, and newsworthiness rewards surprise. Surprising findings are disproportionately statistical flukes. The chapter is explicit that this is structural rather than a failure of individual journalists.

9. A controlled feeding trial measuring LDL cholesterol over 12 weeks is limited primarily because:

  • a) It cannot be randomized
  • b) It measures a surrogate marker rather than a disease outcome
  • c) The sample is always too small to analyze
  • d) Participants cannot be blinded to their own diet
Answer **b.** These trials are often rigorously controlled — sometimes to the gram — but a marker moving for twelve weeks is not the same as an event avoided in thirty years. (d) is true of most diet trials but is not the *primary* limitation here.

10. Which of the following did the chapter list as part of the stable, non-controversial core of nutrition knowledge?

  • a) Optimal macronutrient ratios for weight loss
  • b) Whether saturated fat should be limited
  • c) Folic acid before and during early pregnancy reduces neural tube defects
  • d) The ideal eating window for metabolic health
Answer **c.** Supported by randomized evidence and not seriously contested. The other three are live arguments — (a) is Chapter 10, (b) is Chapter 9, (d) is Chapter 21.

11. IARC classifies alcohol as a Group 1 carcinogen. The chapter notes that Group 1 describes:

  • a) The magnitude of cancer risk
  • b) The strength of evidence that the agent causes cancer
  • c) The dose at which cancer occurs
  • d) The number of cancers the agent causes annually
Answer **b.** Group 1 is a statement of *confidence in causation*, not of *how much risk*. Conflating the two is why "alcohol is in the same category as asbestos" is both true and routinely misused. Chapter 12 unpacks this properly.

12. Theo's story is used in the chapter primarily to establish that:

  • a) Most people don't try hard enough to eat well
  • b) Confusion is the predictable result of being careful in a hostile information environment
  • c) Keto is ineffective
  • d) People should stop reading nutrition research
Answer **b.** Theo read good sources and got worse outcomes. The chapter's argument is that the environment punished carefulness — which is why the fix is a method, not more effort.

True / False

Give a one-line justification for each.

13. Because industry funds a substantial share of nutrition research, conclusions supported by many independent lines of evidence across different countries should also be discarded.

Answer **False.** Funding effects are real and are a reason to scrutinize any *single* study. But independent replication across countries with different agricultural industries — and industry influence that pushes in mutually incompatible directions — means the strongest conclusions don't rest on any one funder's work.

14. The chapter concludes that most apparent reversals in nutrition are reversals in narration rather than in evidence.

Answer **True.** That's §1.1's central claim, illustrated with eggs — while acknowledging (§1.7) that genuine evidence reversals do occur and are a sign of a healthy field.

15. Identifying which myth "shape" a claim uses tells you whether the claim is false.

Answer **False.** It tells you why the claim is *spreading*, which is a separate question from truth. It should raise your evidentiary bar — the claim has a tailwind unrelated to accuracy — but a well-shaped claim can still be true.

16. A study being randomized and placebo-controlled is sufficient to establish that its finding is trustworthy.

Answer **False.** Design quality is necessary, not sufficient. Funding source, outcome selection (marker vs. disease), duration, sample representativeness, publication bias, and whether the result replicates all still matter. Chapter 2 formalizes this.

17. The chapter argues that nobody in the five-step pipeline needs to lie in order for a distortion to occur.

Answer **True.** Each participant only needs to be slightly more confident than the previous one, because every stage's incentives reward confidence and punish hedging. Five slight amplifications compound.

Short Answer

18. In two or three sentences, explain the apparent contradiction between "nutrition research is riddled with measurement error and confounding" and "there is a stable body of settled nutrition knowledge," and resolve it.

Answer Both are true because they describe **different tiers of the evidence**. The weaknesses — FFQ error, confounding, small effect sizes — apply most severely to *single observational studies of specific nutrients*, which is exactly where headlines come from. The settled core rests on convergence: multiple study designs, multiple populations, plausible mechanism, and often randomized evidence pointing the same way. Uncertainty about whether a specific nutrient shifts risk by 8% or 12% is entirely compatible with certainty that severe deficiency causes disease and that overall dietary patterns matter.

19. Name the four questions from the Chapter Summary that you should ask of any nutrition claim, and explain why the fourth one must be applied in every direction.

Answer *Compared to what? In whom? How was diet measured? Who benefits if I believe this?* The fourth must be applied symmetrically because selective application is the most common failure of nutrition skepticism: people ask "who profits?" about mainstream guidance and forget to ask it about the supplement seller, the influencer with the discount code, or the author of the contrarian book — all of whom also profit. Applied in one direction only, "follow the money" stops being a tool and becomes a rationalization.

20. Explain why the chapter argues that a book in which every verdict is ❌ would be a failure.

Answer Because uniform rejection is not skepticism but contrarianism — credulity pointed the other way. A functioning evidence framework must be capable of returning *yes*; if it never does, it isn't evaluating anything, it's just applying a prior. The chapter names several claims popular in wellness culture that the evidence supports (creatine, vitamin D in the deficient, fiber, higher protein for older adults), and the six-level verdict scale exists specifically so that "well supported," "unclear," and "untested" are available answers rather than rhetorical concessions.

Applied Scenario

21. Your sister sends you a video. A physician in a white coat, with 800,000 followers, says: "New research shows this common food additive damages the gut lining. I've stopped eating it and so should you. Link in bio for my gut-repair protocol."

Write out, in order, the specific steps you would take to evaluate this — and say what you would tell your sister if you could not find the underlying study at all.

Answer A reasonable sequence: 1. **Who is speaking, and what do they sell?** "Link in bio for my protocol" answers this immediately — there's a product downstream of the claim. Not disqualifying, but it sets the bar. 2. **Is a study cited?** If not, that's close to the end of the analysis. If yes, find it. 3. **If found: what species, what design, what dose, what outcome?** "Damages the gut lining" strongly suggests either cell-culture or rodent work, plausibly at doses far above human intake, measuring a marker of permeability rather than a disease. 4. **Compared to what?** If people stop eating this additive, what replaces the foods it's in? 5. **Where on the pipeline is this version?** A white coat plus a product link plus "new research" is step 4 behavior. **If the study can't be found:** say so plainly. "I couldn't find the research he's referring to" is a legitimate and complete answer — the burden of proof sits with the person making the claim and selling the protocol, not with you. Notice the chapter's framing, though: your sister isn't foolish for sending it. She sent it because she cares about you, and leading with contempt teaches her to stop asking.

Scoring

Score Reading
18–21 You've got it. Move to Chapter 2, which will formalize most of this.
14–17 Solid. Reread §1.4 (the pipeline) and §1.7 (reversals) before continuing.
10–13 Reread §1.2 and §1.3 — the structural and measurement problems are load-bearing for the whole book.
Under 10 Reread the chapter, then do Exercises A1–A10. Don't skip ahead; Chapter 2 assumes this material.

Regardless of score: if you have not yet written your Belief Inventory (the Project Checkpoint), do that before reading Chapter 2. It genuinely stops working once you know what's coming.